Negative Sequence Relaying Applications in Ungrounded and High Impedance Grounded Industrial Systems

David Beach (Basler Electric Company) Abstract--Industrial Power systems that serve critical
process loads are often served by high impedance grounded, or ungrounded, power distribution systems. These systems are intended to allow continued operation of the process loads despite a single phase-to-ground fault on the system, allowing time for maintenance personnel to find and resolve the fault without unplanned equipment shutdown. The risk associated with these systems is that a second phase-to-ground fault, on a different phase, creates a phase-to-phase fault with much higher fault currents than those associated with the initial phase-to-ground fault. These phase-to-phase faults can take a variety of paths, often through process equipment not intended for carrying fault current. The fault current is limited by the impedance of the fault path and may not be sufficient to quickly trip larger phase current-sensing circuit breakers. Where the faults occur with sufficient electrical distance between the two connections to ground, it is possible for two ground-fault trip units to see the fault and respond. When there are not ground fault trip units or both fault points occur downstream of a single ground fault trip unit, other means are necessary to ensure proper fault clearing. This paper will explore the use of negative sequence overcurrent sensing to detect these faults at levels and speeds not achievable using conventional circuit breaker technology. Potential fault conditions will be explored, with comparisons between phase overcurrent, ground overcurrent, and negative sequence sensing and the levels at which each might be able to detect and interrupt the fault current. The paper will also show that the use of negative sequence sensing makes possible the detection and interruption of these fault currents securely and at values less than maximum load currents. Finally, the paper will provide guidelines to aid in determining which systems, or where in the system, the maximum benefit can be achieved through use of negative sequence sensing.

The presence of a second ground fault before the first fault is found and repaired creates a line-to-line fault through ground. The amount of ground current returning to the source remains at the low levels of a single fault; but the fault current between the two faults can, depending on fault resistance, be much higher, and is likely to flow through paths not intended to carry fault currents. Depending on the fault impedance and the rating of the downstream device(s) nearest the fault locations, the fault can persist for extended periods. While the fault persists, equipment is subject to voltage rise, heating due to the current flow, and the effects of arcing gaps and other impedance points. Some attempts are made to use residual ground-fault sensing elements to detect these phase-to-phase faults. When the entire fault current flows through a single residual ground-fault device, the fault will not be detected as the ground-fault device will measure only the current returning to the source through ground. The remaining fault current flows out on one phase conductor and returns on another phase conductor, the sum being very close to zero. When the fault occurs between two circuits with different residual ground-fault sensing elements, the fault will be seen and will trip one or both devices provided the fault current is sufficiently large. II. NEGATIVE SEQUENCE When using symmetrical components to analyze unbalanced system operation and fault conditions, the negative sequence is generated by unbalance. A balanced three-phase load or a bolted three-phase fault will be entirely positive sequence. Unbalanced loads or asymmetric faults will reduce the proportion of positive sequence current and will produce negative sequence currents. These unbalanced conditions will also produce zero sequence currents to the extent that load or fault current flows in some conductor other than the threephase conductors. In the ungrounded or high impedance systems, line-to-neutral loads are not permitted, eliminating one significant source of potential negative sequence currents in the system. The presence of phase-to-phase loads will produce a negative sequence component to the normal load currents but, in many cases, this negative sequence current will have a very low magnitude relative to the total current or to the positive sequence current. Negative sequence currents on the secondary of transformers will be reflected in the primary currents. Where delta-wye transformers are used, the negative sequence current in the primary will be shifted 30 degrees from the secondary, but in the opposite direction to the shift of the positive sequence currents. When there are multiple step-down transformers connected to the system under consideration, the negative

I. BACKGROUND In the interest of maintaining continuous process operation, many industrial systems are operated either ungrounded or high-impedance grounded. In these systems, a single line-toground fault will draw currents small enough that circuit breakers or fuses used as feeder or branch circuit protection will not trip. This allows the faulted circuit to remain in service while the ground fault is located and corrective measures planned and implemented. These single faults on the system are typically detected by a shift in voltage-to-ground of each phase. The detector may be a voltage relay or may be as simple as light bulbs connected in grounded wye. Currentbased detection of single ground faults in ungrounded systems is not practical while, in high impedance grounded systems, current in the grounding impedance indicates the presence of a fault but not the location. Some high-impedance grounding systems provide a means to pulse the ground current between two values, such as 5A and 10A, to aid in tracing the ground fault.

sequence currents from each will tend to cancel to the extent that the unbalances behind each transformer are on different phases. Three-phase motor loads, the principal load on many systems of this type, do not produce negative sequence currents during normal operation. These characteristics suggest that negative sequence currents may provide a means of identifying and clearing phase-tophase faults in these systems. III. ANALYSIS SINGLE FAULTS This is the classic line-to-line-to-ground (LLG) fault. For the Phase B to Phase C fault analyzed, the sequence diagram is as shown in Fig. 1. The impedances shown as Z1c, Z2c, and Z0c are the impedances in the common branch of the circuit as used with the parallel faults shown in Fig. 3. In this case, there are no branch currents but, to keep the notation similar, the common branch notation is shared between the diagrams.

base, the 55.4 ohms of resistance necessary to limit ground faults to 5A equates to a 3*Zg of over 720 p.u., while the total positive sequence and negative sequence impedances are less than 0.1 p.u. It can be seen that essentially no current flows in the zero sequence branch and that the positive and negative sequence currents are of the same magnitude and opposite angle. The very low current in the zero sequence branch of the network shows why ground elements cannot be used to detect second ground faults in this configuration. A ground element set to detect a LLG fault would also detect a single line-toground fault (SLG). Since the goal is to allow SLG faults to persist until repaired, the inability to distinguish LLG and SLG faults precludes using sensitive ground elements in this application. The negative sequence branch of the network carries essentially the same current as the positive sequence branch. With this configuration, the positive and negative sequence currents each have a magnitude approximately equal to the faulted phase current divided by the square root of 3. The presence of this negative sequence current can be used to distinguish these faults from load for fault currents less than the maximum load current. The extent to which this can be less than maximum load current is discussed below under Setting the Negative Sequence Elements. IV. ANALYSIS PARALLEL FAULTS The sequence network for this fault is shown in Fig. 2. This is not a common sequence network configuration, but portions of it are suggested in several of the references, particularly Power System Protection and Analysis of Faulted Power Systems.

Fig. 1. Sequence diagram for Phase B to Phase C fault analyzed

The fault resistance is shown as Zf/2 rather than Zf to align with the fault impedance given in the study results. The study fault impedance is the total impedance between the two faulted phases while the typical derivation of the sequence network has two equal impedances, both labeled Zf, from each phase to the fault location. For the system considered, 3*Zg is very large compared to the other impedances. On a 1 MVA
Fig. 2. Sequence network for fault

The impedances Z1c, Z2c, and Z0c in the network are the common portion, between the transformer (Zxf) and the point the two branches split. The impedances Z1a, Z2a, and Z0a are in one branch to the fault while impedances Z1b, Z2b, and Z0b are in the other branch. For the purposes of this analysis, it is assumed that half of the total fault impedance, Zf, is on each side of ground. Given the magnitude of 3*Zg relative to the rest of the impedances in the system, the distribution of the total fault impedance between the phases and ground has little effect on the results. The diagram shows the sequence network for two simultaneous SLG faults. The phase-shifting transformers are a result of the symmetry change in creating a SLG fault in a phase other than Phase A. In this case, branch a has a fault on Phase B, while branch b has a fault on Phase C. If one of the faults were on Phase A, all three of the transformers for that portion of the network would be 1:1 with no phase shifting. If the 3*Zg impedance is considered to be infinite, it can be seen that the zero sequence currents in the two branches would be equal in magnitude and 180 degrees apart in phase. With the zero sequence currents in the branches of equal magnitude, it becomes evident that the negative and positive sequence currents in the two branches are also of the same magnitude as the zero sequence currents. In the case under consideration, the amount of current in the common portion of the zero sequence circuit is very small compared to the zero sequence current in the two branches, and the currents in the branches are very nearly 180 degrees out of phase. The phase shifting in the positive and negative sequence networks cause the two branch currents to be 60 degrees out of phase with each other. The result of this is that the current in the common portion of the network has a magnitude 3 times larger than currents in the branches, with the phase angle in the common portion 30 degrees from the two branches. If the impedances in the branches are combined and placed in the common portion of the network, this network will give the same results as the simpler network of Fig. 2. For a fault represented by the network of Fig. 3, breakers or relays on the branches will see positive, negative, and zero sequence currents, each with a magnitude equal to 1/3 of the fault current. Breakers or relays on the common portion of the network will see only positive and negative sequence currents, and these currents will each have a magnitude equal to the fault current divided

grounded through a resistor to limit phase-to-ground fault currents to 5A. A main switchboard is connected to the transformer through 50 feet of 9 parallel sets of 500 kcmil copper cables. At the service switchboard, the three-phase fault current available is 36,854A.
Source

P S

Service XF

Service Fdr 3200/3 Main Service Board 1600/3 Bkr 1 1600A Fdr 1 BUS A 800/3 Bkr A 800A Fdr A-B BUS B 400A Bkr B 400A Fdr B-C BUS C 200A Bkr C 200A Fdr C-D BUS D 1600/3 Bkr 2 1600A Fdr 2 BUS E 800/3 Bkr E 800A Fdr E-F BUS F 400A Bkr F 400A Fdr F-G BUS G 200A Bkr G 200A Fdr G-H BUS H

Fig. 3. Simple system designed to allow testing various fault location on feeders of many common sizes

3.

The internal feeders are all 50 feet in length; the 200A, 400A, and 800A feeders are made of one, two, and four sets of 3/0 copper conductors in parallel. The 1600A feeders are five sets of 500 kcmil copper in parallel. The breakers at the service switchboard are ANSI rated Low Voltage Power Circuit Breakers. Beyond that board, the 800A breakers are equipped with solid-state trip units and ground fault sensing. The 400A and 200A breakers have thermalmagnetic trips. The system uses circuit breakers from two of the major manufacturers, although this appears to have a limited effect on the results. 3

V. THE STUDY SYSTEM The results presented in this paper are based on a simple system, Fig. 3, designed to allow testing various fault locations on feeders of many common sizes. The system starts with a source supplying 5500 Amperes of fault current at 12.47kV to a 2500kVA transformer with an impedance of 5.75%. On the 480V side of the transformer, the neutral is

The breaker phase and ground settings were selectively coordinated to the extent possible; some overlap in the instantaneous region of the phase settings is unavoidable. Negative sequence trip elements are those of one line of relays. The negative sequence lines on the trip curves are the relay sense and operate times; an additional 0.12 seconds was added to each for breaker operating time. Coordination curves are shown in Figs. 4 through 9 at the end of the paper. VI. THE STUDY Results of the study calculations are shown in Table I. Faults of increasing resistance were created at each load bus and between load busses of the same rating of the different branches. For each fault resistance, phase currents and sequence currents were calculated using a MathCAD sheet created for the purpose; these values were checked against results from a commercial fault-calculation software package. With these values available, breaker trip times were calculated for tripping based on phase, ground, and negative sequence values. For many fault resistances, it was found that one or more of the elements would not cause a breaker to trip. When the entire fault happens downstream of a single ground unit, the residual current seen by the ground unit is limited by the grounding impedance, and the ground unit will not trip for these faults. This will be discussed further, under Single Faults. When the two fault locations are downstream of different ground units, there is residual current seen by the ground unit. The magnitude of this residual current is not limited by the grounding impedance but is limited by the circuit and fault impedances. As the fault resistance increases, fault currents decrease and eventually are low enough that protective elements will no longer pick up. The clearing times listed in the data include clearing times in seconds, N/A where the listed element would never trip, and >nnn when the element would pick up but the clearing time is over 1000 seconds or tripping depends on where in the tolerance band the specific breaker falls. With tripping times available, energy dissipated in the fault path is calculated. For tripping times of N/A or >nnn, a clearing time of 1000 seconds was chosen arbitrarily on the assumption that someone would eventually notice the problem and turn off enough circuits to clear one of the phase-toground faults. The energy dissipated in each fault, if cleared by the phase, ground, or negative sequence elements, are listed and the lowest value shown. Based on these energy levels, a best element is selected. The final column looks at those cases where the negative sequence element is the best element and compares the energy dissipated prior to the negative sequence element clearing the 4

fault with the energy dissipated if the second best element clears the fault. VI. RESULTS Refer to Table I. As anticipated, ground elements in the common portion of the system do not see, and will not clear, these LLG faults. A somewhat unexpected result was that when there are ground elements in the branches, those ground elements are often the most effective element in clearing the fault. The results for tests 54-60 and 80-81 show the ground element clearing these faults faster than either the phase or negative sequence elements; however, in these cases the ground element is one or more breakers upstream from the fault location. This is indicated in the Table by the highlighted area. For bolted faults, phase elements operating in their instantaneous region will provide faster fault clearing than either negative sequence or ground elements (see tests 16, 23, 30, 37, 45, 53, 61, 69, and 77). Tests 1, 6, and 11 provide similar results with a phase element in its short time region. Where a ground element is not available or will not see the fault, and where the fault resistance is high enough that the phase elements do not trip in the fastest portion of their trip characteristics, the negative sequence elements can be seen to reduce significantly the amount of energy released during fault events. In most cases, the energy released in the fault is less than 1% of that released were the negative sequence element not present. In many cases, the energy released is less than 0.1% of that otherwise released. VII. CONSIDERATIONS IN APPLYING NEGATIVE SEQUENCE ELEMENTS It is important to establish the normal level of negative sequence current in the system to be protected. If the entire load is connected between two phases, with no load on the third phase, the negative sequence current would be the maximum load current divided by 3 . Setting the negative sequence pickup at this level will ensure that the negative sequence element will not pick up for any load condition, but it may be set much higher than need be. For a system where 75% or more of the load is 3-phase motor load, it may be possible to set the negative sequence elements at 25% of maximum load, the setting used in this study. A negative sequence element can overreach a downstream phase sensing element. To check for possible overreach, the negative sequence element should be plotted at 3 times its actual current and compared to the phase element. For example, to check a negative sequence element with a 100A pickup (I2) against a downstream 125A breaker (phase current), plot the negative sequence element as though it had a 173A pickup (100A * 3 ). Because this overreach does not occur on balanced load, it may be decided to allow overreach

for increased sensitivity similar to the way ground elements typically overreach downstream phase elements. Because these faults are through a path not designed to carry fault currents, there is a possibility of arcing during the fault condition. If a 50T type of element is used, there is the possibility of the timing resetting during the arcing condition if the current drops sufficiently. Using a 51 element and selecting an integrating (or electromechanical) reset characteristic will allow the element to resume timing following intermittent interruptions. Selecting the flattest curve available in the relay will allow relatively uniform timings for higher currents while maintaining coordination at lower current levels. The curves used in this study are a shorttime inverse curve family based on the IAC-55 relay curves and a definite time curve family based on the CO-6 relay. These two curve families are similar with the CO-6 type curves fitting above the IAC-55 type curves. The lower two curves are the IAC-55 curves and the upper two curves are the CO-6 curves. It is also possible that an existing system that has never had negative sequence sensing will have bad power factor correction capacitors or starter contactors that do not make or break uniformly. One system was found to have several problems creating undesired negative sequence currents in excess of 50% of load. Once these problems were tracked down and repaired, the negative sequence current on the system was reduced to less than 10% of maximum load current. VIII. APPLICATION AT HIGHER VOLTAGES While this study is based on a 480V system, the findings are applicable at higher voltages. Medium-voltage industrial systems with relays would have different curves for phase and ground elements, but the general relationships remain. In general, relay ground elements can be set lower than the ground elements of 480V breakers, but relay ground elements are also unable to detect these faults when both faulted phases are downstream of the ground element. IX. CONCLUSIONS Phase protection is effective for bolted faults and faults with very low resistance but can be very slow for faults with more than minimal resistance; sufficient resistance and a phase element will not clear the fault. Ground protection can be effective if the ground element sees only part of the fault. Ground elements are blind to LLG faults on these systems when the entire fault is downstream of the ground element. Negative sequence elements can detect LLG faults on ungrounded or high-impedance grounded systems whether seeing part of the fault or the entire fault. If the negative sequence element sees the entire fault it will respond if the fault current exceeds

sequence element sees only one branch of the fault it will respond if the fault current exceeds 3 times the pickup.

3 times pickup. If the negative

5

CURRENT IN AMPERES 1000

100

3200/3 Main - Phase 1600A Bkr 1 - Phase 800A Bkr A - Phase 400A Bkr B - Phase

10 200A Bkr C - Phase

TIME IN SECONDS

0.10

0.01 0.5 1

10

100

1K

10K

ABCD Phase.tcc Ref. Voltage: 480 Current in Amps: x 10 1Line001.drw

Fig. 4.

CURRENT IN AMPERES 1000

3200/3 Main - Ground 100

1600A Bkr 1 - Ground 10 800A Bkr A - Ground

TIME IN SECONDS

400A Bkr B - Phase 200A Bkr C - Phase 1

0.10

0.01 0.5 1

10

100

1K

10K

ABCD Ground.tcc Ref. Voltage: 480 Current in Amps: x 10 1Line001.drw

Fig. 5.

CURRENT IN AMPERES 1000 3200/3 Main 1600A Bkr 1 800A Bkr A 400A Bkr B 200A Bkr C

100

3200/3 Main - Neg Seq 10 1600A Bkr 1 - Neg Seq

TIME IN SECONDS

800A Bkr A - Neg Seq 400A Bkr B - Neg Seq 1 200A Bkr C - Neg Seq

0.10

0.01 0.5 1

10

100

1K

10K

ABCD Neg Seq.tcc Ref. Voltage: 480 Current in Amps: x 10 1Line001.dr

Fig. 6.

CURRENT IN AMPERES 1000

3200/3 Main - Phase 100 1600A Bkr 2 - Phase 800A Bkr E - Phase 400A Bkr F - Phase 10 200A Bkr G - Phase

TIME IN SECONDS

0.10

0.01 0.5 1

10

100

1K

10K

EFGH Phase.tcc Ref. Voltage: 480 Current in Amps: x 10 1Line001.drw

Fig. 7.

CURRENT IN AMPERES 1000

100

3200/3 Main - Ground 10 1600A Bkr 2 - Ground

TIME IN SECONDS

800A Bkr E - Ground 400A Bkr F - Phase 1 200A Bkr G - Phase

0.10

0.01 0.5 1

10

100

1K

10K

EFGH Ground.tcc Ref. Voltage: 480 Current in Amps: x 10 1Line001.drw

Fig. 8.

10

CURRENT IN AMPERES 1000 3200/3 Main 1600A Bkr 2 800A Bkr E 400A Bkr F 200A Bkr G

100

3200/3 Main - Neg Seq 10 1600A Bkr 2 - Neg Seq

TIME IN SECONDS

800A Bkr E - Neg Seq 400A Bkr F - Neg Seq 1 200A Bkr G - Neg Seq

0.10

0.01 0.5 1

10

100

1K

10K

EFGH Neg Seq.tcc Ref. Voltage: 480 Current in Amps: x 10 1Line001.dr

Fig. 9.

11

3I0

Neg Seq % Energy 0.26% 0.08% 0.12% 0.34% 0.07% 0.08% 0.12% 0.34%

0.33% 0.20% 0.05% 0.06% 0.08% 0.13% 0.27% 0.16% 0.08% 0.06% 0.08% 0.13%

Test Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Pos 16994 1093 831 551 368 16994 1093 831 551 368 9492 629 479 318 212 15266 1084 826 549 367 276 221 15266 1084 826 549 367 276 221 8054 621 474 315 211 159 127 12507 1066 816 544 365 275 220 184 12507 1066 816 544 365 275 220 184 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 28476 1887 1437 954 636 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 24162 1863 1422 945 633 477 381 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 B Phase 0.22 260 N/A N/A N/A 0.14 >300 N/A N/A N/A 0.22 260 N/A N/A N/A 0.026 133 225 >353 N/A N/A N/A 0.04 161 276 644 N/A N/A N/A 0.026 133 230 >360 N/A N/A N/A 0.011 61.7 118 317 >320 >885 N/A N/A 0.017 74 130 390 >295 >746 N/A N/A Zero 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 9492 629 479 318 212 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 8054 621 474 315 211 159 127 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 Lowest Energy 191 609 540 531 1026 121 609 540 531 1026 114 178 136 91 304327 18 387 304 230 187 189 230 28 387 304 230 187 189 230 15 87 67 45 45 36 46 5 247 197 138 99 79 69 62 8 247 197 138 99 79 69 62

Faulted Bus B Phase C Phase Bus A Bus A Bus A Bus A Bus A Bus A Bus A Bus A Bus A Bus A Bus E Bus E Bus E Bus E Bus E Bus E Bus E Bus E Bus E Bus E Bus A Bus E Bus A Bus E Bus A Bus E Bus A Bus E Bus A Bus E Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus B Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus F Bus B Bus F Bus B Bus F Bus B Bus F Bus B Bus F Bus B Bus F Bus B Bus F Bus B Bus F Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus C Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G Bus G

Fault Ohms 0.001 0.25 0.33 0.5 0.75 0.001 0.25 0.33 0.5 0.75 0.001 0.25 0.33 0.5 0.75 0.001 0.25 0.33 0.5 0.75 1 1.25 0.001 0.25 0.33 0.5 0.75 1 1.25 0.001 0.25 0.33 0.5 0.75 1 1.25 0.001 0.25 0.33 0.5 0.75 1 1.25 1.5 0.001 0.25 0.33 0.5 0.75 1 1.25 1.5

Fault Amps B Phase C Phase 29436 29433 1892 1892 1439 1439 953 953 637 637 29436 29433 1892 1892 1439 1439 953 953 637 637 28476 28474 1888 1888 1437 1437 953 952 637 637 26442 26440 1876 1876 1430 1430 950 949 635 635 477 477 382 382 26442 26440 1876 1876 1430 1430 950 949 635 635 477 477 382 382 24163 24161 1862 1862 1422 1422 946 946 634 634 477 477 382 382 21644 21662 1846 1846 1412 1412 942 942 632 632 476 475 381 381 318 318 21644 21662 1846 1846 1412 1412 942 942 632 632 476 475 381 381 318 318

Sequence Amps Neg 16994 1092 830 550 367 16994 1092 830 550 367 9492 629 479 318 212 15265 1083 825 548 366 275 220 15265 1083 825 548 366 275 220 8054 621 474 315 211 159 127 12507 1065 815 543 364 274 220 183 21507 1065 815 543 364 274 220 183

Clearing Time (sec) C Phase B Ground C Ground Neg. Seq. 0.22 N/A N/A 0.54 260 N/A N/A 0.68 N/A N/A N/A 0.79 N/A N/A N/A 1.17 N/A N/A N/A 3.37 0.14 N/A N/A 0.54 >300 N/A N/A 0.68 N/A N/A N/A 0.79 N/A N/A N/A 1.17 N/A N/A N/A 3.37 0.14 0.25 0.2 0.54 >300 0.25 0.2 0.99 N/A 0.25 0.2 1.48 N/A 0.25 0.2 N/A N/A N/A N/A N/A 0.026 N/A N/A 0.41 133 N/A N/A 0.44 225 N/A N/A 0.45 >353 N/A N/A 0.51 N/A N/A N/A 0.62 N/A N/A N/A 0.83 N/A N/A N/A 1.26 0.04 N/A N/A 0.41 161 N/A N/A 0.44 276 N/A N/A 0.45 644 N/A N/A 0.51 N/A N/A N/A 0.62 N/A N/A N/A 0.83 N/A N/A N/A 1.26 0.04 0.15 0.1 0.41 165 0.15 0.1 0.49 285 0.15 0.1 0.54 650 0.15 0.1 0.7 N/A 0.15 0.16 1.43 N/A 0.16 0.28 N/A N/A 0.25 0.45 N/A 0.011 N/A N/A 0.28 61.7 N/A N/A 0.29 118 N/A N/A 0.3 317 N/A N/A 0.31 >320 N/A N/A 0.33 >885 N/A N/A 0.35 N/A N/A N/A 0.38 N/A N/A N/A 0.41 0.017 N/A N/A 0.28 74 N/A N/A 0.29 130 N/A N/A 0.3 390 N/A N/A 0.31 >295 N/A N/A 0.33 >746 N/A N/A 0.35 N/A N/A N/A 0.38 N/A N/A N/A 0.41

Energy Released (kJoules) Phase Ground Neg. Seq. 191 866390 468 232678 894916 609 683338 683338 540 454105 454105 531 304327 304327 1026 121 866390 468 894916 894916 609 683338 683338 540 454105 454105 531 304327 304327 1026 114 162 438 231695 178 882 681440 136 1009 453628 91 453628 304327 304327 304327 18 699126 287 117019 879844 387 151834 674817 304 450775 450775 230 302419 302419 187 227529 227529 189 182405 182405 230 28 699126 287 141655 879844 387 186249 674817 304 290299 450775 230 302419 302419 187 227529 227529 189 182405 182405 230 15 58 239 115279 87 425 153476 67 360 290848 45 313 301467 45 431 227529 36 227529 182405 46 182405 5 468852 131 52564 851929 247 77636 657936 197 140647 443682 138 299568 299568 99 226100 226100 79 181451 181451 69 151686 151686 62 8 468852 131 63043 851929 247 85532 657936 197 173036 443682 138 299568 299568 99 226100 226100 79 181451 181451 69 151686 151686 62

Best Element Phase Neg Seq Neg Seq Neg Seq Neg Seq Phase Neg Seq Neg Seq Neg Seq Neg Seq Phase Ground Ground Ground None Phase Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Phase Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Phase Ground Ground Ground Ground Ground Ground Phase Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Phase Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq

0.47% 0.25% 0.10% 0.03% 0.03% 0.04% 0.04% 0.39% 0.23% 0.08% 0.03% 0.03% 0.04% 0.04%

Table I.

12

3I0

Neg Seq % Energy

2.84% 1.70% 0.76% 0.28% 0.12% 0.06% 0.03%

0.21% 0.10% 0.05% 0.03% 0.02%

Test Number 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 Pos 6007 604 464 311 209 158 127 106 8890 1033 796 536 361 273 219 183 8890 1033 796 536 361 273 219 183 3864 573 446 303 206 156 125 105 18021 1812 1392 933 627 474 381 318 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 2.49 11592 1719 1338 909 618 468 375 315 B Phase 0.012 62.8 123 322 >320 >320 N/A N/A 0.018 5.6 9.4 21.2 57 130 268 594 0.017 0.019 0.027 76 154 307 633 >200 0.015 6.1 10.3 22.1 60 134 291 605 Zero 6007 604 464 311 209 158 127 106 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 0.83 3864 573 446 303 206 156 125 105
Lowest Energy 4 82 64 44 44 34 45 53 4 127 100 69 47 36 29 24 4 15 17 69 47 36 29 24 2 14 17 41 43 35 29 25

Faulted Bus B Phase C Phase Bus C Bus G Bus C Bus G Bus C Bus G Bus C Bus G Bus C Bus G Bus C Bus G Bus C Bus G Bus C Bus G Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus D Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus H Bus D Bus H Bus D Bus H Bus D Bus H Bus D Bus H Bus D Bus H Bus D Bus H Bus D Bus H Bus D Bus H
Energy Released (kJoules) Phase Ground Neg. Seq. 4 32 91 51549 82 246 78706 64 198 140149 44 148 295788 44 115 223729 34 103 180500 45 103 150734 53 125 4 242456 39 4476 799236 127 5899 627542 100 9109 429665 69 16699 292969 47 28962 222784 36 48120 179551 29 88972 149784 24 4 242456 39 15 799236 127 17 627542 100 32655 429665 69 45117 292969 47 68395 222784 36 113656 179551 29 149784 149784 24 2 13 21 14 74 118 17 59 94 9110 41 66 17131 43 46 29224 35 36 51426 46 29 89476 52 25

Fault Ohms 0.001 0.25 0.33 0.5 0.75 1 1.25 1.5 0.001 0.25 0.33 0.5 0.75 1 1.25 1.5 0.001 0.25 0.33 0.5 0.75 1 1.25 1.5 0.001 0.25 0.33 0.5 0.75 1 1.25 1.5

Fault Amps B Phase C Phase 18022 18019 1812 1812 1393 1392 933 933 628 628 473 473 380 380 317 317 15572 15570 1788 1788 1379 1379 927 927 625 625 472 472 379 379 316 316 15572 15570 1788 1788 1379 1379 927 927 625 625 472 472 379 379 316 316 11593 11591 1718 1718 1337 1337 908 908 617 617 467 467 376 376 314 314

Sequence Amps Neg 6007 604 464 311 209 158 127 106 8890 1032 796 535 361 272 218 182 8890 1032 796 535 361 272 218 182 3864 573 446 303 206 156 125 105

Clearing Time (sec) C Phase B Ground C Ground Neg. Seq. 0.017 0.15 0.1 0.28 77 0.15 0.1 0.3 133 0.15 0.1 0.31 419 0.15 0.1 0.34 >295 0.15 0.16 0.39 >746 0.15 0.29 0.46 N/A 0.25 0.44 0.57 N/A 0.35 0.63 0.83 0.018 N/A N/A 0.159 5.6 N/A N/A 0.159 9.4 N/A N/A 0.159 21.2 N/A N/A 0.16 57 N/A N/A 0.16 130 N/A N/A 0.161 268 N/A N/A 0.162 594 N/A N/A 0.162 0.017 N/A N/A 0.159 0.019 N/A N/A 0.159 0.027 N/A N/A 0.159 76 N/A N/A 0.16 154 N/A N/A 0.16 307 N/A N/A 0.161 633 N/A N/A 0.162 >200 N/A N/A 0.162 0.017 0.15 0.1 0.159 0.019 0.15 0.1 0.16 0.028 0.15 0.1 0.16 79 0.15 0.1 0.161 160 0.15 0.17 0.162 332 0.16 0.29 0.164 667 0.26 0.45 0.166 >210 0.35 0.66 0.168

Best Element Phase Ground Ground Ground Ground Ground Ground Ground Phase Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Phase Phase Phase Neg Seq Neg Seq Neg Seq Neg Seq Neg Seq Phase Phase Phase Ground Ground Ground Neg Seq Neg Seq

63.04% 48.08%

Table I, (cont.)

13

X. REFERENCES ----; Electrical Transmission and Distribution Reference Book. 1997, ABB Power T&D Company, Raleigh. ----; Power System Protection Volume 1: Principles and components. Edited by The Electricity Training Association. 1995 Electricity Association Services Limited, The Institution of Electrical Engineers, Stevenage, United Kingdom. Anderson, Paul M.; Analysis of Faulted Power Systems. 1995 The Institute of Electrical and Electronics Engineers. Wiley-Interscience, New York. Blackburn, J. Lewis; Protective Relaying: Principles and Applications. 1998, Marcel Dekker, New York. Clarke, Edith; Circuit Analysis of A-C Power Systems. 1943, General Electric Company, John Wiley & Sons, New York. Grainger, John J. and Stevenson, William D. Jr.; Power System Analysis. 1994, McGraw-Hill Book Co. Singapore.

XI. BIOGRAPHY David Beach received a BS degree in Electrical Engineering from California State University, Fresno in December of 1982. Since that time, he has become a Registered Professional Engineer, licensed in the states of California, Oregon, and Washington. David worked in the Consulting Engineering business until February 2005 when he joined Basler Electric Company as a Senior Application Engineer. Beach is a Senior Member of the IEEE, a member of the Industrial Applications Society and the Power Engineering Society of IEEE, and represents Basler on the work groups extending IEEE Standard 1547.

14

If you have any questions or need additional information, please contact

P .A.E. Les Pins, 67319 Wasselonne Cedex FRANCE Tel +33 3.88.87.1010 Fax +33 3.88.87.0808 e-mail: franceinfo@basler.com No. 59 Heshun Road Loufeng District (N), Suzhou Industrial Park, 215122, Suzhou, P .R.China Tel +86(0)512 8227 2888 Fax +86(0)512 8227 2887 e-mail: chinainfo@basler.com 55 Ubi Avenue 1 #03-05 Singapore 408935 Tel +65 68.44.6445 Fax +65 65.68.44.8902 e-mail: singaporeinfo@basler.com

Basler Electric Company

Route 143, Box 269, Highland, Illinois U.S.A. 62249 Tel +1 618.654.2341 Fax +1 618.654.2351 e-mail: info@basler.com